7815
J. Am. Chem. SOC.1994,116, 7815-7826
Conversion of CH4 to CH3OH: Reactions of COO+with CH4 and D2, Co+ with CH30D and D20, and Co+(CH30D) with Xe Yu-Min Chen, D. E. Clemmer,' and P. B. Armentrout' Contribution from the Department of Chemistry, University of Utah, Salt Lake City, Utah 841 12 Received January 18, 1994. Revised Manuscript Received April 25, 1994'
-
Abstract: The mechanisms and energetics involved in the conversion of CHI to CH3OH by COO+are examined by using guided ion beam mass spectrometry. The forward and reverse reactions, COO++ CH4 Co+ + CHgOH, the collisional activation of Co+(CH,OH), and the related reactions, COO++ D2 Co+ + D20, are studied. It is found that the oxidations of methane and D2 by COO+,both exothermic reactions, do not occur until overcoming activation barriers of 0.56 f 0.08 and 0.75 f 0.04 eV, respectively. The behavior of the forward and reverse reactions in both systems is consistent with reactions that proceed via the insertion intermediates R-Co+-OH, where R = CH3 or H. The barrier is probably attributable to a four-centered transition state associated with addition of RH across the COO+bond. In the Co+ + CHJOH system (where CH30D labeled reactant is used), reactions explained by initial C-H and 0-H activation are also observed. The reaction mechanisms and potential energy surfaces for these systems are derived and discussed. Phase space theory calculations are used to help verify these details for the COO+ + DZ system. Thermochemistry for several species including CoOH+,COD+,CoH, CoCH3+, Co+(CH3OD), CoOCH3+,and possibly OCoCH3+ is derived from measurements of reaction thresholds.
-
Introduction The direct conversion of methane to methanol has been studied for more than a half century because of its great economic and scientific importance.' Although this oxidation reaction, CH4
+ '/202 CH,OH -*
AH298= -1.32 eV
(1)
is thermodynamically favored? the best catalyst at present provides a yield of only about 8%.3 Catalytic conversion of CH4 to CH3OH has become an active research area with the search for a more efficient catalyst listed as one of ten challenges for catalysis.4 One means of providing fundamental information regarding this process is to study a prototypical gas-phase reaction
MO+ + CH,
-
CH,OH
+ M+
+
(2)
t
+
Present address: Department of Chemistry, Northwestern University,
2145 Sheridan, Evanston, IL 60208.
-
COO'
Such studies can potentially provide quantitative information regarding the thermodynamics and mechanisms for this process, while simultaneously examining the periodic trends in the chemistry. At present, only a few gas-phase studies of reaction 2 have been made.5-6 Schrijder and Schwarz reported a Fourier transform ion cyclotronresonance (FTICR) study of the reaction of FeO+ with CH4.5 At thermal energies, the reaction produces 57% of FeOH+ CH3,2% of FeCH2+ HzO, and 41% of Fe+ CHsOH, which were explained by a reaction mechanism involving a CH3-Fe+-OH intermediate. Additionalexperimental
+
and theoretical studies of various possible intermediates were later conducted as we11.6 In a recent paper on the reactions of ScO+, TiO+, and VO+ with D2,' we suggested that there should be two main criteria in evaluating the efficiency of reaction 2. First, reaction 2 should be exothermicor thermoneutral. Because the reaction, 0 CH4 CHsOH, has an exothermicity of -3.846 i 0.005 eV, Table 1,reaction 2 will beendothermicforanyMO+ witha bondstronger than 3.85 eV, such as ScO+,TiO+, or VO+, but no other first-row transition-metalion o ~ i d e Second, . ~ ~ ~ MO+ shouldhave a suitable electron configuration such that reaction 2 conserves spin. As discussed elsewhere,' COO+is a promising candidate for efficient conversion of CH4 to CH30H. The 0 K bond energy of Co+-0 is 3.25 f 0.05 eV,9J0so reaction 3 is thermodynamically favored.
*Abstract published in Advance ACS Abstracts, June 15, 1994. (1) For a review, see: Gesser, H. D.; Hunter, N. R.; Prakash, C. B. Chem. Rev. 1988,85, 235. (2) Lias, S.G.; Bartmess, J. E.; Liebman, J. F.; Holmes, J. L.; Levin, R. D.; Mallard, W. G. J. Phys. Chem. Ref. Data 1988, 17, Supp. 1 (GIANT tables). ( 3 ) Chem. Eng. News 1993, May 10, 22. (4) Chem. Eng. News 1993, May 31,27. (5) Schrbder, D.; Schwarz, H. Angew. Chem., Int. Ed. Engl. 1990, 29, 1433. Schwarz, H. Angew. Chem., Int. Ed. Engl. 1991,30, 820. (6) Schrbder, D.; Fiedler, A,; HruSdk, J.; Schwarz, H. J. Am. Chem. Soc. 1992, 114, 1215.
+
+ CH,
-
Co+
+ CH,OH
AHo = -0.60 f 0.05 eV (3)
Further, it was suggested that COO+ has a suitable electron configurationto activate CH4, and that reaction 3 is spin allowed? In this paper, we examine the energetics and mechanism of reaction 3 in both the forward and reverse directions, as well as probing how one of its intermediates, Co+(CH30H),decomposes when activated by collisions with Xe. We also study the related but simpler interaction of Coo+with Dz and its reverse, a reaction that has been briefly studied at thermal energies by Schrijder et al.1' We are able to determine a fairly detailed picture of the potential energy surfaces of these reactions and find in both systems that a reaction barrier restricts the oxidation process and its reverse. These conclusions lead to a more careful assessment (7) Clemmer, D. E.; Aristov, N.; Armentrout, P. B. J . Phys. Chem. 1993, 97. 544. (8) Armentrout, P. B.; Clemmer, D. E. In Energetics o/Organometallic Species, Sim&s, J. A. M., Ed.; Kluwer: Netherlands, 1992; p 321. (9) Armentrout, P. B.; Kickel, B. L. In Organometallic Ion Chemistry; Freiser, B. S., Ed.; in press. (10) Fisher, E. R.; Elkind, J. L.; Clemmer, D. E.; Georgiadis, R.; Loh. S. K.; Aristov, N.; Sunderlin, L. S.;Armentrout, P. B. J. Chem. Phys. 1990,93, 2676. The value cited here is an average value reassessed in ref 9. (1 1) Schrbder, D.; Fiedler, A.; Ryan, R. F.; Schwarz, H. J . Phys. Chem. 1994, 98, 68.
0002-7863/94/1516-7815$04.50/00 1994 American Chemical Society
Chen et al.
7816 J . Am. Chem. SOC.,Vol. 116, No. 17. 1994 Table 1. Bond Energies and Enthalpies of Formation at 0 K species 0 H D OH OD
ArHo (eV) 2.558(0.001)a 2.239O 2.278O 0.405(0.003)e 0.382(0.001)c
bond Co+-O CO+-H Co+-D Co-H Co+-OH Co+-CH2 Co+-CH3 Co+-H20
DO(eV) 3.25(0.05)b 1.98(0.06)e 2.01 (0.06): 2.05(0.05)" 1.86(0.05)/ 1.89(0.06)" 3.13(0.04),d3.12(0.13),8
3.08(0.1 3)h 3.29(0.05y 2.10(0.04),' 2.10(0.08)" 1.67(0.06),"' 1.61(0.13),g 1.74(0.2)"
CH2 CH3 H20
4.02(0.03)' 1.553(0.004)' -2.476(0.001)'
HDO D20 CH2O CH2OH CH3O CH3OH CHlOD CHI
-2.513(0.001)" -2.552(0.001)" -1.086(0.005)0S -0.12(0.01)q 23.0(0.3)$ 21.8' 0.25(0.04)& Co+-OCH3 -1.976(0.003)0' OCO+-CH3 22.32(0.05)" -2.005(0.003)' Co+-CHsOD 1.53(0.08)" -0.688(0.004)"~
a Chase, M. W .;Davies, C. A.; Downey, J. R.; Frurip, D. J.; McDonald, R. A.; Syverud, A. N. J. Phys. Chem. Ref. Data 1985, 14, Suppl. No. 1 (JANAF Tables). References 10 and 9. C Reference 36. This work. Gurvich, L. V.; Veyts, I. V.; Alcock, C. B. Thermodynamic Properties of Indioidual Substances, 4th ed.; Hemisphere: New York, 1989; Vol. 1, Part 2.fReference 38. #Reference 28. No temperature specified. h Reference 29. No temperature specified. Lwpold, D. G.; Murray, K. K.; Stevens Miller, A. E.;Lineberger, W. C. J. Chem. Phys. 1985,83, 4849. Reference 16. Berkowitz, J.; Ellison, G. B.; Gutman, D. J . Phys. Chem., in press. Reference 35. Reference 39. * 298 K value from ref 44. 0ArH2g8 value of Pedley, J. B. Naylor, R. D.; Kirby, S. P. Thermochemical Data of Organic Compounds, 2nd ed.; Chapman and Hall: New York, 1986. p Adjusted to 0 K by using information in footnote a. q Average of values in ref 37 and footnote k. 'Derived from results in ref 33. * Adjusted to 0 K by using information in footnote t . ' Value taken from the following and adjusted for ArH(CH3OH) used here: Chen, S. S.;Wilhoit, R. C.; Zwolinski, B. J. J. Phys. Chem. Ref. Data 1977, 6, 105.
and D2 were used directly without further purification, and CHsOD and D2O were purified by several freeze-pump-thaw cycles with liquid N2 to eliminate noncondensible impurities before use. Ion Source. Cobalt ions are produced in a dc-discharge flow tube (DC/FT) source.13J5 The flow gases used in this experiment are He and Ar, maintained at pressures of -0.65 and -0.06 Torr, respectively. A dc discharge at a voltage of -2.2 kV is used to ionize argon and then accelerate these ions into a cobalt metal cathode to create Co+ ions. The ions are then swept down a meter-long flow tube by the He and Ar flow gases and undergo 10s collisions with the flow gases. Trace amounts of high-lying excited states (>2.1 eV) can survive these flow conditions, but they are easily removed by introducing 0 2 to the flow tube several centimeters downstream at a pressure of -2 mTorr. We believe these conditions produce Co+ ions in their ground electronic state. All results reported here are consistent with this, as are tests involving a number of other reactions.16 COO+ was produced in the flow tube by introducing N20 at a pressure of less than 1 mTorr, together with He and Ar flowing through the dcdischarge source. The reaction of Co+ with N20 to form COO+ is exothermic by about 1.6 eV, but it is observed to be inefficient at low collision energies." Co+(CHsOD) was produced in the flow tube by three-body stabilization of Co+ with CH3OD, introduced at a pressure of less than 1 mTorr at the midpoint of the flow tube. The COO+and Co+(CH3OD) ions are cooled by lo5 collisions with the flow gases. Collision-induced dissociation (CID) of COO+ with Xe is consistent with ions that are not internally excited. We assume that these ions are in their ground electronic states and that the internal energy of these clusters is well described by a Maxwell-Boltzmann distribution of rotational and vibrational states corresponding to 298 K. Previous work from this I a b o r a t ~ r y ~ ~has J " ~shown ~ that these assumptions are valid. Data Analysis. Cross sections are modeled by using eq 4,"
-
-
(4)
where E is the relative kinetic energy of the ions, EOis the 0 K reaction threshold, uoisan energy-independentscaling factor, and n is an adjustable parameter. Equation 4 takes into account the thermal internal energy of the reactants by treating the calculated cross section as a sum over of the ground-state electron configuration for COO+and its vibrational states (with energies E, and populations g,) as described previously,ls and by including the thermal rotational energy. In this suitability for activation of CH4. study, the vibrational energies of D2, CH,, and COO+ at 298 K are negligible; only the vibrational energies of CH3OD and Co+(CH,OD) Experimental Section were included in eq 4. The vibrational frequencies of CHsOD were taken General Procedures. The guided ion beam instrument on which these from S h i m a n ~ u c h i . ~No ~ vibrational frequencies are available for experiments were performed has been described in detail p r e v i ~ u s l y . l ~ * ~ ~ Co+(CHsOD), so we estimated these frequencies as the vibrational Ions are created in a flow tube source, described below. The ions are frequencies of CHoOD plus three cobalt-methanol vibrational frequencies extracted from the source, accelerated, and focused into a magnetic sector (one stretching and two bending) of 330, 192, and 106 cm-Ie26 The total momentum analyzer for mass analysis. Mass-selected ions are slowed rotational energy of the reactants at 298 K is Emc= 3 k ~ T / 2= 0.039 eV to a desired kinetic energy and focused into an rf octopole ion guide that for the Co+ + D20, Co+ + CH,OD, and Co+(CHpOH) Xe systems, radially traps the ions.14 The octopole passes through a static gas cell = 0.053 eV for the COO+ + D2 system, and 5 k ~ T / 2= 0.066 eV 2k~T containing the neutral reactant. Gas pressures in the cell are kept low for the COO+ CH4 system. Before comparison with the data, eq4 must (between 0.08 and 0.3 mTorr) so that multiple ion-molecule collisions beconvolutedover thereactant neutral and ion kinetic energy distributions, are improbable. All results reported here are due to single bimolecular as described previously.12 encounters, as verified by pressure dependent studies. Product and (15) Schultz, R. H.; Crellin, K. C.; Armentrout, P.B. J. Am. Chem. Soc. unreacted beam ions are contained in the guide until they drift out of the 1991, 113, 8590. gas cell where they are focused into a quadrupole mass filter for mass (16) Haynes, C. L.; Armentrout, P. B. Organometallics, accepted for analysis and then detected. Ion intensities are converted to absolute publication. cross sections as described previously.1* Uncertainties in cross sections (17) Armentrout, P.B.; Halle, L. F.; Beauchamp, J. L. J. Chem. Phys. are estimated to be *20%. 1982, 76,2449. (18) Schultz, R. H.; Armentrout, P.B. J. Chem. Phys. 1992, 96, 1046. Laboratory ion energies (lab) are converted to energies in the center(19) Khan, F. A.; Clemmer, D. C.; Schultz, R. H.; Armentrout, P. B. J . of-mass frame (CM) by using the formula ECM= E w n / ( m + M), where Phys. Chem. 1993,97,7978. M and m are the ion and neutral reactant m a w , respectively. The (20) Fisher, E. R.; Kickel, B. L.; Armentrout. P.B. J. Phys. Chem. 1993, absolute zero and distribution of the ion kinetic energy are determined 97,10204. by using the octopole beam guide as a retarding potential analyzer.12The (21) Fisher, E. R.; Kickel, B. L.; Armentrout, P.B. J. Chem. Phys. 1992, 97, 4859. uncertaintyin theabsoluteenergyscaleis*0.05 eV (lab). Thedistribution (22) Dalleska, N. F.; Honma, K.; Armentrout, P.B. J. Am. Chem. Soc. of ion energies is nearly Gaussian and has an average fwhm of about 0.35 1993,115, 12125. eV (lab). Unless otherwise stated, all energies cited below are in the CM (23) Chen, Y.-M.; Armentrout, P.B. C k m . Phys. Lett. 1993,210, 123. frame. (24) Armentrout, P.B. In Advances in Gas Phase Ion Chemistry; Adams, CHsOD, D2, and D20 are obtained from Cambridge Isotope N. G.; Babcock, L. M.,Eds.; JAI: Greenwich, 1992; Vol. 1, p 83. Laboratories with purity stated as 99%,99.8%, and 99.9%, respectively, (25) Shimanouchi, T. Tables of Molecular Vibrational Frequencies Consolidated; NSRDS-NBS No. 39, 1972; Vol. 1. and CHI is obtained from Matheson with a purity stated as 99.99%. CH4 (26) These three frequencies were estimated from a theoretical calculation for Mg+(CH,OH): Sodupe, M.; Bauschlicher, C. W .Chem.Phys. Lett. 1992, (12) Ervin, K. M.; Armentrout, P. B. J. Chem. Phys. 1985,83, 166. 195,494. The stretching mode at 415 cm-l was scaled to 330 cm-1 by using (13) Schultz,R. H.; Armentrout, P.B. I n t . J. MassSpectrom. Ionprocesses 1991. -107. 29. a Morse potential model, and the two bending mode frequencies were used without modification. Variation of these frequencies by *30% has nosignificant (14) Teloy, E.; Gerlich, D. Chem. Phys. 1974, 4, 417. Gerlich, D. effect on the threshold analysis. Diplomarbeit, University of Freiburg, Federal Republic of Germany, 1971.
+
+
----. .
J . Am. Chem. SOC.,Vol. 116, No. 17, 1994 7817
Conversion of CH4 to CHjOH ENERGY (eV.
0.0
10.0
Lob) 20.0
+
30.0
restrict the formation of the CoOD+ D channel (which has a reduced mass of 1.96 amu, much smaller than that of the reactants, 14.95amu) relative to theCoD+ + ODchannel(wherethereduced mass of 13.90 amu is comparable to that of the reactants). This is an effect that we have discussed in detail previously.3' Analysis of the cross section data for reactions 5 and 6 with eq 4 yields the optimum parameters listed in Table 2. The thresholds measured are in excellent agreement with the thermodynamic values, showing that there are no reaction barriers in excess of the reaction endothermicities for either reaction. Despite a careful search, no evidence of COO+formation was observed in this study, indicating that its cross section is less than 0.01 A,. This is somewhat surprisingbecause the thermodynamic threshold for the dehydrogenation of water by Co+ is 1.86 f 0.05 eV, lower than those of reactions 5 and 6. This negative observation can be further investigated by studying the reverse reaction. COO+ Dz. Reaction of translationally excited COO+with Dz leads to three ionic products, CoOD+, Co+, and COD+,formed in reactions 7, 8, and 9, respectively.
+
2.0
6:O 6.'0 CM) Figure 1. Cross sections for reactions of Co+ with DzO as a function of 0.0
4:O ENERGY
